We need some way of accessing the FlyEM dataset without having to download 14TB or so to our personal computers. Ideally the way we host it will be such that Windows, MacOS, and Linux can all mount the network location, so that we can work with the dataset as if it was local to our machines (as opposed to calling RESTful APIs or other such methods).
Ultimately we want the data to remain in a state similar to how it would be right out of an electron microscope. If we want to host the data on an SQL database instead, the process of putting the data into the database needs to be automated.
Before storing the data in a database we must first ensure tiles and layers are aligned properly (as such, the first steps of STS remain the same). How ever we store the data, it must retain information such as the coordinate range for a given tile. The coordinates will be X,Y,Z (with Z being the layer).
We need some way of accessing the FlyEM dataset without having to download 14TB or so to our personal computers. Ideally the way we host it will be such that Windows, MacOS, and Linux can all mount the network location, so that we can work with the dataset as if it was local to our machines (as opposed to calling RESTful APIs or other such methods).
Ultimately we want the data to remain in a state similar to how it would be right out of an electron microscope. If we want to host the data on an SQL database instead, the process of putting the data into the database needs to be automated.
Before storing the data in a database we must first ensure tiles and layers are aligned properly (as such, the first steps of STS remain the same). How ever we store the data, it must retain information such as the coordinate range for a given tile. The coordinates will be X,Y,Z (with Z being the layer).